Pull the numbers from the tools that already hold them, assemble them in one place while nobody is working, and deliver one message before the week starts. That part is a few hours of build. The part that decides whether anyone reads it is choosing which four lines belong in the report, and cutting the rest.
Almost every team already produces a weekly report. It is made by hand on Friday afternoon by someone who would rather be doing something else, and it is read properly by two people. Automating it is worth doing. Automating it without changing what is in it is not.
01Friday afternoon is spent gathering, not thinking
Watch the hour that goes into the report and you will not see analysis. You will see someone opening four tabs, copying a figure out of the CRM, another out of the project tool, a third out of the spreadsheet where invoices are tracked, and pasting them into a document that looked exactly the same last week. That is transcription, not reporting.
It also lands at the worst possible moment. End of week, everyone tired, and the numbers get a glance instead of a read. The gathering squeezes out the thinking, week after week, and nobody ever decides to fix it because each individual Friday it only costs an hour. We put a price on that kind of repeated copying in what re-typing data actually costs, and the arithmetic is uncomfortable.
02Decide the four lines before you build anything
If the honest answer is no, it does not belong in the report. Run the question down your current one and most reports lose half their content. That is the point. A report longer than a screen is a report nobody finishes.
What usually survives falls into four groups, and four is generally enough.
Every line carries a number and a direction. A figure without last week's figure next to it is trivia, and trivia is what people stop reading.
03What the build actually looks like
A scenario runs on Sunday evening. It queries each source in turn, the CRM for what came in, the project tool for what moved, the invoicing tool or the sheet for what went out, and writes one row into a single spreadsheet that keeps the history. Then it formats a short message and posts it on Monday morning in the channel where the team already is, rather than in a document somebody has to remember to open.
The history matters more than the message. After ten weeks that sheet is the only place in the company where the same four numbers sit side by side over time, and that is where the useful conversations start. The Monday message is the reminder. The sheet is the asset.
Nothing here is exotic. It is the same three-part shape as every other scenario, described in ten Make scenarios for small businesses: something happens on a schedule, checks run, actions follow.
04Why most automated reports get ignored
- It arrives at the wrong time. A report that lands Friday at six is read on Monday, if at all. Monday at seven is read before the first meeting, because there is still time to act on it.
- It has no direction. Fourteen new leads means nothing. Fourteen against nine last week means something, and takes one extra column to produce.
- It lives in a document. Every click between the person and the number loses readers. Put the four lines in the message itself and link to the detail for the two people who want it.
- Nobody owns it. A report addressed to everyone is read by no one. One name at the bottom, whose job is to say one sentence about the stuck line, changes the whole thing.
All four failures have the same fix. Shorter, comparative, delivered where the team already looks, with a name attached.
05When you should not build this
- The report has never existed. Automating a habit you do not have yet produces a message nobody was waiting for. Send it by hand for a month first. If people ask for it when it is late, it is worth building.
- Everything is already in one tool. If all four numbers live in the same system, use that system's own scheduled view. Building a scenario to move data out of a tool and back into a message from the same tool is work for nothing.
- The sources disagree. If the CRM and the spreadsheet hold two different revenue figures, automating the report will publish that disagreement every Monday with your name on it. Reconcile first, automate second.
- The numbers are not stable yet. A definition of "closed" that changes every month means a history you cannot compare. Freeze the definitions, then build.
If you want to know whether this is even the task worth starting with, the arithmetic takes ten minutes and it is in the 10-minute self-audit. And if it turns out to be worth building, you own it: accounts in your name, the scenario named and commented, the sheet in your drive. That is how we work, and it is the reason the report keeps running after we leave.
Common questions, honest answers.
How do you automate a weekly report +
A scheduled scenario queries each source, the CRM, the project tool and wherever invoices are tracked, writes one row into a spreadsheet that keeps the history, then posts a short message in the team channel on Monday morning. The technical part is a few hours. The decision that matters is which four lines go in it and which ones you cut.
What should a weekly business report contain +
Four lines are usually enough: what came in, what went out, what is stuck, and what is coming. Each one with a number and last week's number beside it, because a figure without a comparison is trivia. If you would not notice a line missing, remove it.
Why does nobody read our weekly report +
Usually one of four reasons. It arrives Friday evening when nobody can act on it, it shows numbers without last week's for comparison, it sits in a document that takes a click to open, or it is addressed to everyone and therefore owned by no one. Shorter, comparative, delivered where the team already looks, with one name attached.
Curious if we can help
Book your free 30-minute AI scan. We'll map where your team actually loses time, and tell you honestly what's worth automating. And what isn't.
FREE AI SCAN